1<!DOCTYPE html> 2<html lang="en"> 3<head> 4 5 <meta charset="utf-8" /> 6 <meta http-equiv="X-UA-Compatible" content="IE=edge" /> 7 8 <title>About</title> 9 <meta name="HandheldFriendly" content="True" /> 10 <meta name="viewport" content="width=device-width, initial-scale=1.0" /> 11 12 <link rel="stylesheet" type="text/css" href="/assets/built/screen.css?v=0aa202968c" /> 13 14 <link rel="canonical" href="https://kvfrans.com/about/" /> 15 <meta name="referrer" content="no-referrer-when-downgrade" /> 16 17 <meta property="og:site_name" content="kevin frans blog" /> 18 <meta property="og:type" content="website" /> 19 <meta property="og:title" content="About" /> 20 <meta property="og:description" content="Hi, I'm Kevin! Welcome to my webpage. I am a PhD student at Berkeley AI Research under Pieter Abbeel and Sergey Levine. Before that I did my undergrad + master's at MIT CSAIL, under Phillip Isola. My interests are in understanding deep learning and scalable self-improving systems. Generally, my prior research" /> 21 <meta property="og:url" content="https://kvfrans.com/about/" /> 22 <meta property="article:published_time" content="2016-08-07T12:39:41.000Z" /> 23 <meta property="article:modified_time" content="2025-01-29T22:17:53.000Z" /> 24 <meta name="twitter:card" content="summary" /> 25 <meta name="twitter:title" content="About" /> 26 <meta name="twitter:description" content="Hi, I'm Kevin! Welcome to my webpage. I am a PhD student at Berkeley AI Research under Pieter Abbeel and Sergey Levine. Before that I did my undergrad + master's at MIT CSAIL, under Phillip Isola. My interests are in understanding deep learning and scalable self-improving systems. Generally, my prior research" /> 27 <meta name="twitter:url" content="https://kvfrans.com/about/" /> 28 <meta name="twitter:label1" content="Written by" /> 29 <meta name="twitter:data1" content="Kevin Frans" /> 30 31
31<script type="application/ld+json"> 32{ 33 "@context": "https://schema.org", 34 "@type": "Article", 35 "publisher": { 36 "@type": "Organization", 37 "name": "kevin frans blog", 38 "url": "https://kvfrans.com/", 39 "logo": { 40 "@type": "ImageObject", 41 "url": "https://kvfrans.com/favicon.ico", 42 "width": 48, 43 "height": 48 44 } 45 }, 46 "author": { 47 "@type": "Person", 48 "name": "Kevin Frans", 49 "url": "https://kvfrans.com/author/kevin-frans/", 50 "sameAs": [] 51 }, 52 "headline": "About", 53 "url": "https://kvfrans.com/about/", 54 "datePublished": "2016-08-07T12:39:41.000Z", 55 "dateModified": "2025-01-29T22:17:53.000Z", 56 "description": "Hi, I'm Kevin! Welcome to my webpage. I am a PhD student at Berkeley AI Research\n[https://bair.berkeley.edu/] under Pieter Abbeel\n[https://people.eecs.berkeley.edu/~pabbeel/] and Sergey Levine\n[https://people.eecs.berkeley.edu/~svlevine/]. Before that I did my undergrad +\nmaster's at MIT CSAIL [https://www.csail.mit.edu/], under Phillip Isola\n[http://web.mit.edu/phillipi/].\n\nMy interests are in understanding deep learning and scalable self-improving\nsystems. Generally, my prior research has been", 57 "mainEntityOfPage": { 58 "@type": "WebPage", 59 "@id": "https://kvfrans.com/" 60 } 61} 62 </script>
62 63 64 <meta name="generator" content="Ghost 3.41" /> 65 <link rel="alternate" type="application/rss+xml" title="kevin frans blog" href="https://kvfrans.com/rss/" /> 66 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/prism/1.28.0/themes/prism.min.css" crossorigin="anonymous" referrerpolicy="no-referrer" /> 67 68<link 69 rel="stylesheet" 70 href="https://cdn.jsdelivr.net/npm/[email protected]/dist/katex.min.css" 71 integrity="sha384-AfEj0r4/OFrOo5t7NnNe46zW/tFgW6x/bCJG8FqQCEo3+Aro6EYUG4+cU+KJWu/X" 72 crossorigin="anonymous" 73/> 74 75</head> 76<body class="page-template page-about"> 77 78 <div class="site-wrapper"> 79 80 81<header class="site-header"> 82 <div class="outer site-nav-main"> 83 <div class="inner"> 84 <nav class="site-nav"> 85 <div class="site-nav-left-wrapper"> 86 <div class="site-nav-left"> 87 <a class="site-nav-logo" href="https://kvfrans.com">kevin frans blog</a> 88 <div class="site-nav-content"> 89 <ul class="nav"> 90 <li class="nav-home"><a href="https://kvfrans.com/">Home</a></li> 91 <li class="nav-about nav-current"><a href="https://kvfrans.com/about/">About</a></li> 92 <li class="nav-publications"><a href="https://kvfrans.com/publications/">Publications</a></li> 93</ul> 94 95 </div> 96 </div> 97 </div> 98 <div class="site-nav-right"> 99 100 </div> 101</nav> 102 </div> 103 </div> 104</header> 105 106 107<main id="site-main" class="site-main outer"> 108 <div class="inner"> 109 110 <article class="post-full post no-image page no-image"> 111 112 <header class="post-full-header"> 113 <h1 class="post-full-title">About</h1> 114 </header> 115 116 117 <section class="post-full-content"> 118 <div class="post-content"> 119 <figure class="kg-card kg-image-card"><img src="https://kvfrans.com/content/images/2022/08/Screen-Shot-2022-08-06-at-8.24.24-PM.png" class="kg-image" alt></figure><!--kg-card-begin: markdown--><p>Hi, I'm Kevin! Welcome to my webpage. I am a PhD student at <a href="https://bair.berkeley.edu/">Berkeley AI Research</a> under <a href="https://people.eecs.berkeley.edu/~pabbeel/">Pieter Abbeel</a> and <a href="https://people.eecs.berkeley.edu/~svlevine/">Sergey Levine</a>. Before that I did my undergrad + master's at <a href="https://www.csail.mit.edu/">MIT CSAIL</a>, under <a href="http://web.mit.edu/phillipi/">Phillip Isola</a>.</p> 120<p>My interests are in understanding deep learning and scalable self-improving systems. Generally, my prior research has been in reinforcement learning and generative modelling. That said, the world is vast, and there are many things worth a study.</p> 121<p>I like to write to think. This blog contains mostly thoughts for my own thinking, but you are free to browse around.</p> 122<p>In the past, I spent a year as a researcher at <a href="https://crosslabs.org">Cross Labs</a> investigating open-ended learning. At <a href="https://sizigistudios.com/">Sizigi</a>, I built AI-powered artistic tools. Before that, I worked on generative models at <a href="https://www.autodesk.com/research">Autodesk Research</a> and was on the reinforcement learning team at <a href="https://openai.com/">OpenAI</a>.</p> 123<p>If you are interested in the topics above, shoot me an email (hi at kvfrans.com)! It helps if you've been brainstorming a specific idea or started some preliminary work; we'll have a starting point to discuss.</p> 124<!--kg-card-end: markdown--><!--kg-card-begin: markdown--><p>Timeline</p> 125<ul> 126<li>2017: <a href="https://openai.com/">OpenAI</a></li> 127<li>2018: Autodesk Research</li> 128<li>2018: Start B.S. + M.Eng at MIT</li> 129<li>2019: Sizigi</li> 130<li>2020-2021: Gap Year @ <a href="https://www.crosslabs.org/">CrossLabs</a></li> 131<li>2021: Started <a href="https://paragraphai.com">ParagraphAI</a></li> 132<li>2018-2023: <a href="http://web.mit.edu/phillipi/">Phillip Isola's lab</a> at MIT CSAIL</li> 133<li>2023-Now: PhD @ <a href="https://bair.berkeley.edu/">Berkeley AI Research</a></li> 134</ul> 135<!--kg-card-end: markdown--> 136 </div> 137 </section> 138 139 </article> 140 141 </div> 142</main> 143 144 145 146 147 <footer class="site-footer outer"> 148 <div class="site-footer-content inner"> 149 <section class="copyright"><a href="https://kvfrans.com">kevin frans blog</a> © 2026</section> 150 <nav class="site-footer-nav"> 151 <a href="https://kvfrans.com">Latest Posts</a> 152 153 154 <a href="https://ghost.org" target="_blank" rel="noopener">Ghost</a> 155 </nav> 156 </div> 157 </footer> 158 159 </div> 160 161 162
162<script 163 src="https://code.jquery.com/jquery-3.4.1.min.js" 164 integrity="sha256-CSXorXvZcTkaix6Yvo6HppcZGetbYMGWSFlBw8HfCJo=" 165 crossorigin="anonymous"> 166 </script>
166 167
167<script src="/assets/built/casper.js?v=0aa202968c"></script>
167 168 169
169<script> 170 // Parse the URL parameter 171 function getParameterByName(name, url) { 172 if (!url) url = window.location.href; 173 name = name.replace(/[\[\]]/g, "\\$&"); 174 var regex = new RegExp("[?&]" + name + "(=([^&#]*)|&|#|$)"), 175 results = regex.exec(url); 176 if (!results) return null; 177 if (!results[2]) return ''; 178 return decodeURIComponent(results[2].replace(/\+/g, " ")); 179 } 180 181 // Give the parameter a variable name 182 var action = getParameterByName('action'); 183 var success = getParameterByName('success'); 184 185 $(document).ready(function () { 186 if (action == 'subscribe' && (success === null || success === 'true')) { 187 $('body').addClass('subscribe-success'); 188 } 189 190 if (action == 'subscribe' && success === 'false') { 191 $('body').addClass('subscribe-failure'); 192 } 193 194 $('.subscribe-notification .subscribe-close-button').click(function () { 195 $('.subscribe-notification').addClass('close'); 196 }); 197 198 // Reset form on opening subscrion overlay 199 $('.subscribe-button').click(function() { 200 $('.subscribe-overlay form').removeClass(); 201 $('.subscribe-email').val(''); 202 }); 203 }); 204 </script>
204 205 206
206<script> 207 $(function() { 208 var $postContent = $(".post-full-content"); 209 $postContent.fitVids(); 210 }); 211</script>
211 212 213 214
214<script>
vendor: 345 bytes, lines 214-220
214 215 (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ 216 (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), 217 m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) 218 })(window,document,'script','https://www.google-analytics.com/analytics.js','ga'); 219 220 ga('create', '
220UA-53446327-4
vendor: 39 bytes, lines 220-222
220', 'auto'); 221 ga('send', 'pageview'); 222
223</script>
vendor: 1 bytes, line 223
223
224<script src="https://cdnjs.cloudflare.com/ajax/libs/prism/1.28.0/components/prism-core.min.js" integrity="sha512-9khQRAUBYEJDCDVP2yw3LRUQvjJ0Pjx0EShmaQjcHa6AXiOv6qHQu9lCAIR8O+/D8FtaCoJ2c0Tf9Xo7hYH01Q==" crossorigin="anonymous" referrerpolicy="no-referrer"></script>
vendor: 1 bytes, line 224
224
225<script src="https://cdnjs.cloudflare.com/ajax/libs/prism/1.28.0/plugins/autoloader/prism-autoloader.min.js" integrity="sha512-fTl/qcO1VgvKtOMApX2PdZzkziyr2stM65GYPLGuYMnuMm1z2JLJG6XVU7C/mR+E7xBUqCivykuhlzfqxXBXbg==" crossorigin="anonymous" referrerpolicy="no-referrer"></script>
225 226 227</body> 228</html>
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.